MODEL KOZIOL-GREEN UNTUK ESTIMASI FUNGSISURVIVAL PADA OBSERVASI TERSENSOR KANAN;KOZIOL-GREEN MODEL FOR SURVIVAL FUNCTIONESTIMATION ON RIGHT-CENSORED OBSERVATIONS

Survival data analysis refers to some statistical methods to analyze time-toevent data. The analysis often discuss about observation unit’s probability of survive, known as survival function. Survival function can be estimated using nonparametric method. Two well-known estimators to estimate survival function are Kaplan-Meier and Nelson-Aalen. Right censored observations usually found in time-to-event data. Kaplan- Meier and Nelson-Aalen estimator assume that the survival function on the censored time equal to survival function on the time before. Therefore, the estimation of survival function on the censored time is less precise. That problem can be solved with alternate survival function estimator. That is using Koziol-Green Model. The advantage of this alternate estimator is the survival function on the censored time not assumed equal to survival function on the time before, but have its own estimation. Also the calculating method for the alternate estimator is much simple. We only need calculate the proportion of uncensored data and the empirical cumulative probability to estimate the survival function